Back

 Industry News Details

 
Interview with Alan Lee, VP Data Science, Kohl's - Speaker at Global AI Conf - SCA - Jan 2019 Posted on : Oct 19 - 2018

We feature speakers at 3rd Annual Global Artificial Intelligence Conference - 2019 Jan 23 - 25 – Santa Clara to catch up and find out what he or she is working on now and what's coming next. This week we're talking to Alan Lee, VP Data Science, Kohl's Topic - "Data Science And Applications At Kohl's Retail"

Interview with Alan Lee

1. Tell us about yourself and your background.
I am the VP of Data Science and Data Science engineering at Kohl's.  I work in the Digital office based in Milpitas in the Bay Area.  I come from a science background and have been interested in math and programming ever since I was a kid.  My job allows me to constantly be on the cutting edge of technology and find ways to leverage it to solve important practical problems. 

2.  What have you been working on recently?
Recently I have been at Kohl's building out a top notch Data Science team focused on using AI and Machine learning to solve the many interesting problems in retail.  These include a broad range of projects including online recommendations and search, customer targeting, forecasting, and many others across the entire retail organization.   

3. Where are we now today in terms of the state of artificial intelligence, and where do you think we’ll go over the next five years?
We are at a very exciting time.  Technology is rapidly catching up to the theoretical benefits of exciting algorithm innovations in the past few decades and beyond.  We are just beginning to see platforms capable of executing these algorithms quickly, efficiently, and on BigData in the past few years.  Even more exciting yet, these platforms are mostly open-source giving a level of unprecedented collaboration in technology where all can benefit.  Spark, Tensorflow, etc are examples of platforms that have enabled algorithms like deep neural nets and other ML techniques at large scale.  In 5 years, we will see continued breakthroughs and the increasing democratization of data science to those who are not just strictly data scientists.   

4. There is a negative perception around AI and even some leading technology folks have come out against it or saying that it’s actually potentially harmful to society. Where are you coming down on those discussions? How do you explain this in a way that maybe has a more positive beneficial impact for society?
This is technically not the fault of technology itself.  We all have a choice on how to apply technology.  Gunpowder can be used as a weapon or a means to speed up construction.  AI is simply not mature enough to override a human's initial design decision in it's current state.  If we are careful in design, the risk is low.   This however may not hold in a few decades.  Until then I am not too worried about a robot rebellion.

5. When you’re hiring, what types of people are you hiring? The job market for traditional programmers, engineers is  very difficult to get into AI space. Are you hiring from that talent pool or is that a different talent pool? In terms of talent, how do you go about ensuring you get the best AI people at your company?
There are general guidelines for a data scientist that we follow.  The perfect data scientist is an expert statistician, expert developer, and shrewd businessperson with unlimited common sense.  The intersection of all of those probably sound a little unicorny.  In general our data scientist come from either a statistics or computer science related background and have proven that they are smart people with the thirst to learn.  A majority have Ph.D.s or Masters from well known universities.  We welcome them and help them grow in the areas where they may not have had enough training.  This helps us create a very well rounded data science team with proven capabilities of implementing great things. 

6. Will progress in AI and robotics take away the majority of jobs currently done by humans? Which jobs are most at risk?
Every technical revolution takes away jobs but creates them as well.  Manual jobs are at the most risk.  At the same time it opens up a very large field of AI development that has been giving a lot of people opportunities.  Ultimately how we fare as a  society as a whole is not simply about supply and demand of labor.  It also hinges on government policies, how compassionate we are to others, and how we as a society are willing to adapt to large changes. 

7. What can AI systems do now?
For most tasks where you can say whether there is a right or wrong answer and know all the inputs that are needed to make a decision, a system can likely be trained to make a prediction and automate the task in the coming years.  AI/ML systems have become very good at these "supervised" tasks.  We are still very immature in regards to true "AI" technology that pop culture has attached to the term.  Consciousness and intelligence is something we are barely beginning to understand.  With the explosion in computing capability, I am hopeful that this will be as rapidly accelerated as machine learning has been in the recent years. 

8. When will AI systems become more intelligent than people?
I don't know if I am convinced that AI systems will rival human ingenuity anytime in the forseeable future.  Yes, we can build systems that are more efficient and better at predicting outcomes than a single person but who built these systems?  Would these systems have been able to discover quantum mechanics, group theory, philosophize about the meaning of existence?  I am of the opinion that intelligence is less of a "calculator" function and more of a "creation" function where knowledge is used to extrapolate to areas that are relatively unclear.

9. You’ve already hired Y number of  people approximately. What would be your pitch to folks out there to join your Organization? Why does your organization matter in the world?
In a nutshell, Kohl's data science knows what they are doing.  We have experts in data science and data science engineering, and have consistently proven our solutions can beat the best solutions in the market when it comes to Advanced Machine learning applications.  We've successfully leverage Random Forests, gradient boosted trees, countless forms of regression, neural networks, unsupervised methods, and even many deep learning models to impact top line revenue and show lift from existing systems.  But we are just getting started with many more applications coming.  The Advanced/ML hype is real and we are proving it here.  The Data Science Engineering infrastructure was also built from the ground up to support Data science research and implementation.  We have fully invested in all the open source technologies like Spark , Tensorflow, python, and R that are driving innovation in this space.  We have no problems processing the massive amount of interesting information that comes through our systems and applying the most cutting edge AI solutions to process and predict outcomes.  Kohl's is an excellent place to start or bring your career to the next level.  

10. What are some of the best takeaways that the attendees can have from your talk?
The talk is really intended to give a high level overview of how Data science functions at Kohl's and to show some sample applications of how advanced data science technologies are very applicable to retail problems.

11. Any closing remarks
None other than the fact that it is an honor to be able to present at this conference.